Clay Shirky on Coase, Collaboration and Here Comes Everybody
Oct 20 2008

Clay Shirky, author of Here Comes Everybody: The Power of Organizing Without Organizations, talks about the economics of organizations with EconTalk host Russ Roberts. The conversation centers on Shirky's book. Topics include Coase on the theory of the firm, the power of sharing information on the internet, the economics of altruism, and the creation of Wikipedia.

Michael Munger on the Nature of the Firm
Mike Munger, of Duke University, talks about why firms exist. If prices and markets work so well (and they do) in steering economic resources, then why does so much economic activity take place within organizations that use command-and-control, top-down, centralized...
Ronald Coase on Externalities, the Firm, and the State of Economics
Nobel Laureate Ronald Coase of the University of Chicago talks with EconTalk host Russ Roberts about his career, the current state of economics, and the Chinese economy. Coase, born in 1910, reflects on his youth, his two great papers, "The...
Explore audio transcript, further reading that will help you delve deeper into this week’s episode, and vigorous conversations in the form of our comments section below.


Oct 20 2008 at 11:26am

I loved this discussion. I especially liked the discussion of home schooling. My wife home schooled our boys for a while and I am amazed at how progressive home schoolers are and considering that many home schoolers started as reactionaries it is even more amazing. So you got these people start out complaining among other things about modern education methods, a lack of rigor in schools, demanding more home work etc. and then they start home schooling and they end up finishing school work in 4 hours a day and teach kids arithmetic at stores Also home schoolers do collaborate all the time one parent will teach math to another patents children in return for teaching art etc. In our area home schoolers have taken over the 4h club. .

I also like that discussion of your bus ride. The change in amount of freedom that children today have compared to when I was a child amazes me. My 15 year old son has never taken a bus ride alone! When I was 10 I used buses to travel al over time.

BTW I can work while listening to EconTalk.

scott clark
Oct 20 2008 at 11:39am

I also work and listen, sometimes i have to rewind, though.

I like how in this one you had to stop the conversation to question Shirky’s assumptions about what constitutes public goods.

Otherwise, Shirky’s work is especially dope.

Oct 20 2008 at 3:59pm

As a college student, my work still consists of mindless physical labor, so as a result it is a perfect opportunity to listen to your podcast.

I agree, Shirky’s comments were interesting, although I would have enjoyed hearing about more than just fickr and wikipedia. Maybe I will have to buy the book.

Bob Kozman
Oct 20 2008 at 5:36pm

Those of us who find ourselves living physically in outposts of intellectual wasteland can find comfort in online communities of intellectual stimulation. The world is at our fingertips, and there can be no excuse for not knowing the answer (or at least several versions of one) to any question.

I recall that Alexis de Tocqueville wrote that a very important part of the experiment in democracy was public newspapers with wide circulation.

“The effect of a newspaper is not only to suggest the same purpose to a great number of persons, but to furnish means for executing in common the designs which they may have singly conceived.” (5-2)

The internet is a global extension of that concept.

Oct 21 2008 at 12:00am

Russ thanks for another great podcast. I also listen to your podcast along with NPR and others through my day at work. Thanks for the good work, and also thanks for your spots on NPR lately, it has been refreshing to hear your prospective on things. Please keep up the good work.

Oct 21 2008 at 1:38pm

Russ, Great podcast.

You should be aware of the growing problems with Wikipedia. The danger is not so much around mistakes, which by definition everyone should want to correct, but around opinion where a group of highly motivated editors can “correct mistakes” insuring that their version of reality is what people read.

A prime example are all articles surrounding Global Warming that are “corrected” by a small band of zealots to suppress data or publications that question the environmentalist view.

One example (there are many)

Kevin Abraims
Oct 21 2008 at 5:49pm

This was a very interesting podcast. Clay Shirkey is someone I’ve seen speak a few times, and he has a number of interesting insights about the interwebs.

I’m pushing 45 and while sometimes I do feel that congitive dissonance he speaks of, I still am active on Faebook and can see things like that being useful once they have better filering options, both on the publishing end and the reading end.

Keep up the good work Russ, love that podcast. And, support the economy!

I Support The Economy – All proceeds go to help the GDP!

[Link fixed and correction comment removed.–Econlib Ed.]

Jonah Houston
Oct 23 2008 at 11:20am

I’m a relatively new listener (less than a year) but I found this to be far and away one of the most interesting conversations you’ve had.

Keep up the great work.

I’d love to hear more about the role of behavioral economics and it’s juxtaposition to neoclassical in the future.

Oct 23 2008 at 3:24pm

Thanks for an interesting podcast. I think the best part of this conversation was the discussion about cognative surplus.

Online communities create the ability to significantly increase specialization beyond what was previously possible. Combine this with the natural interest many poeple have in learning and creating IP for its own sake, and we now have the potential for incredible progress. Not to mention the decreased cost of informal education.

Take this site for instance, which relies on voluntary contributions of data from cyclists:

That compilation of data would never have existed previously. Take any hobby or profession and there are equivalent examples.

Comments are closed.


About this week's guest:

About ideas and people mentioned in this podcast:Books:


Web Pages:

Podcasts and Blogs:



Podcast Episode Highlights
0:36Intro. Kling podcast: thanks for feedback. Timely stuff: slim operation, taping in advance, editing, changing. Clay Shirky. Role of technology and allowing spontaneous organizations to emerge. Why do they emerge and what stops them? Famous observation, Coase, 1930s: organizations exist to lower transactions costs. If markets are such a good idea, why do we have firms? If everything were a market, the cost of discovering opportunity and writing a contract for every little thing would swamp the ability of the organization to get anything done. Pencils for desk. Markets exist where level of transaction is large enough and complexity of interacting actors large enough that you can't have an organized firm. Communism. Internet mobile phones applications: finding things easier because of search engines; transactions easier. Easier to find people interested in what you are interested in or who have skills complementary to yours. Sharing, collaboration, social tools. What we didn't notice: In additional to nominal Coasian ceiling, also a Coasian floor. There is a set of transactions that no one engages in because you have to be organized to engage in them but no organization is willing to eat the costs. Transactions costs are too high. Middle management is the requirement.
4:45Overhead of maintaining an institution that can lower those costs itself entails costs. Example: 2005, Coney Island, run-down amusement park, Mermaid Parade. On Flickr, photo-sharing, thousands of pictures of the Mermaid Parade. Flickr had added tagging. Automatically got this aggregate view. Cost of doing this institutionally would have included advertising and doing something to coordinate and extract those people's motives. No profit motive; no one would stand up on floor of House of Representatives. What Flickr did was something no institution could do: side-effect. Gap between a bunch of people want to look at this a little bit and bunch of people are willing to pay a lot of money to look at this. There is value there, but not enough to overcome the transactions costs of putting that produce together. Completely unintended. Beyond Flickr's staff's capabilities. Beta program that allows people to aggregate photos and create a 3-D image. Photosynth. Need multiple exposures of a single space to knit together. Average city block, need amateurs involved. Aggregating step is the missing piece. Coasian floor has dropped. Words becoming less important, images becoming more important. Flickr, over 3 billion images; Facebook over 5 million. Most totally uninteresting. Big change here is the fusing of the old pattern of a broadcast network and a communications network. TV different from phone. Now all coming into the same box. Cognitive dissonance. Photo of someone else's niece is totally uninteresting; but the people who do care are very interested. Things are in public but not for the public. Something doesn't have to be interesting to more than two people to be worth putting up on line. Creates incentive to create more pictures. Plus, looking at pictures of other people's families teaches how to take better pictures, greater interest in photography. Old model: had to join a photo club, subscribe to a magazine--high threshold. Gap is now a spectrum.
14:28Economics of organizations: birthday paradox, choosing a movie. What happens to social complexity as the size of the network goes up? Standing in line with 35 other people. Guy next to you says "I'll bet you $100 bucks that 2 people in the line have the same birthday." Most people won't take it because they think there's only a 1 in 10 chance. They are thinking about the wrong thing. Doesn't count individuals. Counts pairs. For three people, three pairs of birthdays; four people, six pairs of birthdays; five, 10 pairs of birthdays. Number of pairs rises with the square of the number of participants, not at the same rate as the number of participants. By the time you get 36 people, over 80% chance that two share a birthday. Nodes. Large organizations are qualitatively different from small ones. Big city is not just a small city on steroids. If you and a friend want to go to a movie, relatively easy. Try to get 10 friends to agree on a movie: 45 comparisons. Soon get to a point where there has to be some formal mechanism like voting, or governance, like "I'm going to this movie." Requirement for unilateralism isn't like it's politically evil; it's the only way to organize large groups in an institutional setting. Different on the web because web privileges strategies of effectiveness over efficiency. Wikipedia: roughly 2 million contributors in English, something like 5 million across multiple languages. Managerial group is tiny by comparison. Much smaller relative to newspaper writers vs. managerial staff. Registering every user, even if it cost only a nickel, would cost a million dollars. Would bankrupt the organization. Wisdom of crowds, Hayek. Sounds more Age of Aquarius: when we see organized behavior we assume someone is organizing it; not old way hierarchies, new way networks.
20:4010-person example: collective action is difficult because of our differences, different preferences, different costs. Democracy: high transaction cost, inefficient? Two kinds of democracy: direct and representative. Direct: Swiss canton system, town meetings, specifically limited by scale. Certain sizes of cantonships that are not allowed to have direct democracy. Ditto in Massachusetts. Sometimes a family has trouble; often done autocratically. Can you do a better job of representing people's preferences in such a way that relies less on representatives. New York City: council may vote to overturn term limits although public has voted for term limits, Mayor wants to run another term and also offering another term to council members. Bug in representative system. Email, mobile phones are a possibility now. Gory action film versus romantic comedy as movie choices, vote produces not will of the group but will of the majority. Where you don't have to have collective action, strange. Live and let live decoupling. Bunch of places where the government has to decide: garbage collection, school. Why? Cheaper to do that way. Flip side of the birthday paradox, Mancur Olson, logic of collective action. Public goods exist in part because if they're to be privately provisioned past a certain scale, everybody has an incentive to opt out. Public goods get undercreated in purely voluntary situations. Problem is that what is defined as a public good seems a stretch. Throughout U.S. history, private schools. Use of the Internet to turbocharge the home schooling movement. Dropped out of state sponsored society but deeply in touch with each other. Infrastructure that could only be provisioned by the state is now being provisioned by families who are networked with one another. Pulling out of the compound versus social component. End of cyberspace: People are using the Internet to augment their real-world activities rather than provide an alternative to them.
28:24Creation of community, new social technique. Pushback, probably from older folks, looking at it saying it's creepy. Low quality, misspelled communications, geeky warped people who do not interact with other human beings. Extreme view. Communication vs. broadcasting; used to spell-checking. People are writing more now. Many criticisms are not directed at the Internet but at the culture of television. Battle fought and lost by the 1960s. Complaints have now shifted to the Internet. Half-time job to watch television, 20-28 hours a week. Idea that the people who are spending that time are now doing some of that with a keyboard and mouse is the calamity: more going on than simply watching TV passively. Ages 25 and below, mobile phones and Internet are ubiquitous; used to socialize with people they know. Danah Boyd, Berkeley sociologists: rise in social networks correlates with reduction in freedom to go outside. Kids are responding by maintaining a social life. Restrained because we are so cautious; teenagers at risk of being victims and perpetrators of criminal activity. Bus in Lexington, MA; subway in Boston to Fenway park as a ten-year old.
33:35Collaboration: History of Wikipedia. Nupedia story: failure of the institutional model of transactions costs to be grafted on the web. Jimmy Wales, 2000, encyclopedia. Larry Sanger, student of philosophy of learning. Seven steps, classic workflow, fact-checking, spell-checking. Seven places where things can grind to a fault. Dozen articles, highly polished. Sanger discovered wikis. Send out little piece of mail. Opposite of world-beating announcement. Took off; by end of the year, Wikipedia was viable. Replaced all of the organization. Success was not to give veto-power to experts. People who are in charge enjoy the fact that if they do nothing, nothing happens. Article on asphalt, used to cover roads, simplistic: once it went up, people started to do better, now two articles, chemical asphalt and asphalt concrete. Ability to switch the process from efficient to effective. Farrago of inefficiency, more people work at Wikipedia than at Wal-mart, world's largest employer. Where is it going? Hard governance questions. Editing of Sarah Palin article. Wikimedia has to figure out balance participation with gaming the system. Same problem with blog comments on Cafe Hayek. Technical question: Clay Shirky's entry, change to garbage facts: what is process by which mistakes get fixed? George Washington page. People put pages on their watch list; they go in and look at the page. Voluntary editors, just people; people who care. Wikis make it easier to undo damage than to do damage. Spray paint, vandalism. WikiTravel--entry on NYC, ridiculous joke entry, same done for Boston. Deleting it took less than 30 seconds, found IP address, found the other things; reset in a few minutes what had taken vandal 2 hours. Reverse vandalism also easy. Edit war happens occasionally. Backstop: keeps evolving till well into the 1990s. Galileo article now is semi-protected, 500 year old flame war. Court of final appeal. Software is on the side of the angels, emboldens the people who want to defend. Ward, original wiki, shifts economics to defend themselves. Everything radically transparent, history list.
47:06Cognitive surplus. Watching TV being a half-time job. Wikipedia and open source software work by aggregating voluntary contributions with leisure time: Where do people find the time to do this? Nobody who works in TV gets to ask that question. Shift from seeing leisure time as an individual problem of what do I do with 20 hours a week to seeing it as an aggregate. Roughly 100 million hours for Wikipedia, big except compared to trillion hours for TV; 100 million hours spent just watching ads. Things like Wikipedia become much more explainable. Surplus is a kind of raw material for the 21st century, couldn't have existed prior to now. If you'd asked economists even as recently as 2000 if Wikipedia could work, they'd have said it couldn't. No glory in it. Wrong; misunderstood the pure pleasure that overcomes some of the free rider problem. First pictures on Flickr, comments, some pleasure from the experience. Massive social public positive externality, unprecedented in human history. Soup kitchen. Wild speculation: Bill Hamilton, biologist: every scientific theory goes through four phases of reaction: that's not true; that's an interesting if perverse way of looking at the world; I see the experimental evidence accounts for certain edge cases; I have always believed it. Behavioral economics has attained that third stage; cases that neoclassical model doesn't work. Haven't moved behavioral model to the center. Wikipedia is described as a positive externality. Theory isn't working hard enough to explain what's going on. Externality: Benefit that I don't get to capture. Wikipedia will stop being viewed as an externality. Social effects are internal to the transaction. Neoclassical economics; altruism; Adam Smith, charity article in CEE. People understood it exists; but models of how much of that would take place were quite sterile. May be beyond the scope of social science. What scientist are willing to regard as both proof and an explanation. Physics, economics more similar to weather than gravity. Probabilistic explanations: cloudy with a chance of collaborations. A lot will be stuck in a world of special cases. Could not create a Wikipedia alternative.
57:30What are we going to do with that cognitive surplus? Standard of living [taping Oct. 10, 2008, stock market not doing too well], bad financial mess now, but if you look back at the last 50 or even 20 years, we neglect what's changed in the workplace. Larry Iannaccone. Today's workplace remarkably safe and less brutal; amount of leisure while you are under the roof of your employer; people listen to these podcasts while at work. Opportunity to create collaborative projects: what might they be? Speculation: greatest successes have come in the domain of creating or sharing intellectual property. Non-technological event: creation of the GNU public license, open source licenses. Injected the right amount of inefficiency into the intellectual property system. Share and share alike. Using copyright law as the vehicle it created more rights for the user and fewer for the creator. Copied by creative commons movement. Explosion of collaborative use. Not an equivalent for collective action. Protest oriented, political protests, short-term high visibility. What's coming: license for collective action, vehicle of incorporation made to work for these kinds of groups. Virtual company in Vermont: group of people who assemble entirely online can be deferred to by the state. State could see that as a competitor. State has regarded businesses as competitors but it's necessary to have that competition. No state; just individual politicians. Great and terrible things in the United States both happened because of Federalism. Profitability vs. power. Vermont could become the place to go.